- 博客(22)
- 收藏
- 关注
RSS订阅
1转载 PYC编译
指定文件夹进行编译PYC，然后将权重文件和图片等都放在 .pyc 文件的当前目录，比较方便运行。参考博客的代码（两个都可以，都尝试过，就分享稍微简单点的大佬代码吧）简单说，PYC文件可以防止新手看源码。
2022-10-06 10:47:12
217
原创 difficult = obj.find(‘difficult‘).text AttributeError: ‘NoneType‘ object has no attribute ‘text‘
VOC转COCO格式标签bug分析
2022-09-15 09:11:03
1006
torch-geometric==1.7.2安装
torch_spline_conv-1.2.1-cp38-cp38-linux_x86_64.whl
torch_sparse-0.6.9-cp38-cp38-linux_x86_64.whl
torch_scatter-2.0.7-cp38-cp38-linux_x86_64.whl
torch_cluster-1.5.9-cp38-cp38-linux_x86_64.whl
然后pip install torch-geometric==1.7.2
2023-01-17
yolov5 6.0的CMakeLists.txt
project(yolov5)
set(OpenCV_DIR "D:\\yolov5_tensorrt\\yolov5_tensorrt\\opencv\\build") #2
set(OpenCV_INCLUDE_DIRS ${OpenCV_DIR}/include) #3
set(OpenCV_LIB_DIRS ${OpenCV_DIR}/x64/vc15/lib) #4
set(OpenCV_Debug_LIBS "opencv_world343d.lib") #5
set(OpenCV_Release_LIBS "opencv_world343.lib") #6
set(TRT_DIR "D:\\yolov5_tensorrt\\yolov5_tensorrt\\TensorRT-7.0.0.11.Windows10.x86_64.cuda-10.2.cudnn7.6\\TensorRT-7.0.0.11") #3
set(TRT_INCLUDE_DIRS ${TRT_DIR}/include) #8
set(TRT_LIB_DIRS ${TRT_DIR}/lib) #9
#
2022-09-29
cudnn7.6.5（对应的cuda10.2）+git+wget
省得大家下载不了，这个是cuda10.2对应的cudnn版本，cuda10.2用迅雷下载也挺快的，也挺大，所以上传不了这里，另外git和wget都是好东西，git可以通过bash，git，sh等指令；wget就是方便大佬们放链接下载用的。详细见XXX
2022-09-17
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人